English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

For autoregressive (AR) modeling of high-resolution images, vectorquantization (VQ) represents an image as a sequence of discrete codes. A shortsequence length is important for an AR model to reduce its computational coststo consider long-range interactions of codes. However, we postulate thatprevious VQ cannot shorten the code sequence and generate high-fidelity imagestogether in terms of the rate-distortion trade-off. In this study, we proposethe two-stage framework, which consists of Residual-Quantized VAE (RQ-VAE) andRQ-Transformer, to effectively generate high-resolution images. Given a fixedcodebook size, RQ-VAE can precisely approximate a feature map of an image andrepresent the image as a stacked map of discrete codes. Then, RQ-Transformerlearns to predict the quantized feature vector at the next position bypredicting the next stack of codes. Thanks to the precise approximation ofRQ-VAE, we can represent a 256$\times$256 image as 8$\times$8 resolution of thefeature map, and RQ-Transformer can efficiently reduce the computational costs.Consequently, our framework outperforms the existing AR models on variousbenchmarks of unconditional and conditional image generation. Our approach alsohas a significantly faster sampling speed than previous AR models to generatehigh-quality images.

Autoregressive Image Generation using Residual Quantization